Beyond Static Testbeds: An Interaction-Centric Agent Simulation Platform for Dynamic Recommender Systems

2025-05-22

AI Summary

Traditional A/B testing incurs high operational costs, while offline evaluation fails to capture the dynamic user-platform interactions inherent in real-world recommendation systems. Existing proxy simulation platforms lack mechanisms for environment plasticity—i.e., adaptive environmental reshaping—thus failing to faithfully model the evolutionary dynamics of recommendation systems. To address this, we propose RecInter, the first agent-based simulation platform for recommender systems that supports environment-plastic interaction. RecInter enables real-time item attribute updates based on user behavior and triggers merchant responses, thereby modeling emergent phenomena—including brand loyalty and the Matthew effect—for the first time in recommendation simulation. The platform integrates multi-dimensional user profiling, a hierarchical agent architecture, chain-of-thought-enhanced fine-tuned LLMs, and a real-time state synchronization engine. Experiments across multiple benchmarks demonstrate significantly improved simulation fidelity and accurate reproduction of real-system evolutionary patterns, establishing RecInter as a high-fidelity, low-cost dynamic evaluation environment for recommender systems.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Search and Optimization: Sampling/Simulation-based Search
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Evaluating and iterating upon recommender systems is crucial, yet traditional A/B testing is resource-intensive, and offline methods struggle with dynamic user-platform interactions. While agent-based simulation is promising, existing platforms often lack a mechanism for user actions to dynamically reshape the environment. To bridge this gap, we introduce RecInter, a novel agent-based simulation platform for recommender systems featuring a robust interaction mechanism. In RecInter platform, simulated user actions (e.g., likes, reviews, purchases) dynamically update item attributes in real-time, and introduced Merchant Agents can reply, fostering a more realistic and evolving ecosystem. High-fidelity simulation is ensured through Multidimensional User Profiling module, Advanced Agent Architecture, and LLM fine-tuned on Chain-of-Thought (CoT) enriched interaction data. Our platform achieves significantly improved simulation credibility and successfully replicates emergent phenomena like Brand Loyalty and the Matthew Effect. Experiments demonstrate that this interaction mechanism is pivotal for simulating realistic system evolution, establishing our platform as a credible testbed for recommender systems research.

Problem

• Simulating dynamic user-platform interactions in recommender systems
• Overcoming limitations of static testbeds for recommender evaluation
• Enhancing realism in agent-based simulation of evolving ecosystems

Innovation

• Dynamic real-time item attribute updates
• Multidimensional User Profiling module
• LLM fine-tuned on CoT interaction data